Low-dose combinations of naloxone and rimonabant produce additive effects on food intake and feeding behaviour, yet abolish the scratching syndrome typically induced by rimonabant per se. To assess the generality of these findings, we have examined the acute effects of low-dose combinations of naloxone (0.1 mg/kg) and the rimonabant derivative AM 251 (0.5 and 1.0 mg/kg) on food intake, feeding behaviour and weight gain in non-deprived male rats. Although ineffective when given alone, combined treatment with naloxone and 0.5 mg/kg AM 251 significantly and selectively suppressed mash intake and time spent feeding. By itself, 1.0 mg/kg AM 251 failed to alter any measure of feeding behaviour but did reduce food consumption and induce scratching behaviour. Co-administration of naloxone with 1.0 mg/kg AM 251 not only significantly suppressed both food intake and feeding behaviour but also simultaneously attenuated AM 251-induced scratching. This profile mirrors earlier findings with naloxone/rimonabant and is consistent with the reported diversity of opioid-cannabinoid system interactions at a more molecular level. Although further studies are required (e.g. 'neutral' CB1 receptor antagonists), current data constitute further proof of concept regarding the anorectic efficacy, selectivity and added value of low-dose polytherapy with opioid and CB1 receptor antagonists.